1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

library(recount3)

# 同 https://jhubiostatistics.shinyapps.io/recount3-study-explorer/
# 可以看到project_home和project，包括TCGA,GTEX和SRA
human_projects <- available_projects()

proj_info <- subset(
  human_projects,
  project == "OV" & project_type == "data_sources"
)

rse_gene_tcga_ov <- create_rse(proj_info)

#counts data
assays(rse_gene_tcga_ov)$counts <- transform_counts(rse_gene_tcga_ov)
# ## Compute TPMs
assays(rse_gene_tcga_ov)$TPM <- recount::getTPM(rse_gene_tcga_ov, length_var = "score")
# ## Check TPM. Should all be equal to 1
colSums(assay(rse_gene_tcga_ov, "TPM")) / 1e6 


# View sample annotation
View(data.frame(colData(rse_gene_tcga_ov)))

# View gene annotation
View(data.frame(exp$tcga.ov.expr@rowRanges))

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14

rse_gene_tcga_ov = recount3::create_rse_manual(
    project = "OV",
    project_home = "data_sources/tcga",
    organism = "human",
    annotation = "gencode_v26",
    type = "gene"
)

#counts data
assays(rse_gene_tcga_ov)$counts <- transform_counts(rse_gene_tcga_ov)
# ## Compute TPMs
assays(rse_gene_tcga_ov)$TPM <- recount::getTPM(rse_gene_tcga_ov, length_var = "score")
# ## Check TPM. Should all be equal to 1
colSums(assay(rse_gene_tcga_ov, "TPM")) / 1e6

1
2
3
4
5
6

kingfisher get -r ERR1739691 -m ena-ascp
# 可以指定列表--run-identifiers-list
# 指定project编号SRP，--bioprojects
# 指定下载方法，--download-methods，可以指定多种，程序会一个方法一个方法的试，包括ena-ascp，ena-ftp，prefetch，aws-http，aws-cp，gcp-cp
# 指定线程数目，--download-threads
# 指定ascp需要的key路径，--ascp-ssh-key

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

zcat input.fastq.gz | bowtie --seedlen 10 -p 48 -v2 -m20 --best -S --strata --chunkmbs 8000 $BOWTIE_IND - output.sam

-M 1           like -m, but reports 1 random hit (MAPQ=0); requires --best
- 为zcat的输入，bowtie不识别gz，所以用zcat管道流入bowtie
-p 48 线程
-v report end-to-end hits w/ <=v mismatches; ignore qualitie
-m20 丢弃超过20次multi-map的reads
--best --strata multi-map的reads只报告最好的比对
--chunkmbs 8000 best-first搜索的最大内存（M）
-S 输出SAM格式
BOWTIE_IND bowtie的索引

1

samtools sort --threads 10 out.sam -o out_sorted.bam --output-fmt BAM && samtools index out_sorted.bam

 1
 2
 3
 4
 5
 6
 7
 8
 9
10

axel -n 10 https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_44/gencode.v44.annotation.gtf.gz

zcat gencode.v44.annotation.gtf.gz | grep "snoRNA" > snoRNA.gtf
zcat gencode.v44.annotation.gtf.gz | grep "Mt_rRNA" > Mt_rRNA.gtf
zcat gencode.v44.annotation.gtf.gz | grep "Mt_tRNA" > Mt_tRNA.gtf
zcat gencode.v44.annotation.gtf.gz | grep "rRNA" > rRNA.gtf
zcat gencode.v44.annotation.gtf.gz | grep "snRNA" > snRNA.gtf
zcat gencode.v44.annotation.gtf.gz | grep "sRNA" > sRNA.gtf
zcat gencode.v44.annotation.gtf.gz | grep "scaRNA" > scaRNA.gtf
zcat gencode.v44.annotation.gtf.gz | grep "miRNA" > miRNA.gtf

1
2
3

# 也来自gencode，tRNA genes predicted by ENSEMBL on the reference chromosomes using tRNAscan-SE
wget https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_44/gencode.v44.tRNAs.gtf.gz
zcat gencode.v44.tRNAs.gtf.gz > tRNA.gtf

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20

# 下载，注意COMPSRA需要JRE
mkdir COMPSRA
cd COMPSRA
wget https://github.com/cougarlj/COMPSRA/raw/master/COMPSRA_V1.0.3.zip
unzip COMPSRA_V1.0.3.zip

# 手动安装star
cd bundle_v1/plug/star
wget -c wget -c https://github.com/alexdobin/STAR/archive/refs/tags/2.5.3a.zip
unzip 2.5.3a.zip


cd ../../..

# 下载注释数据，如果下载的文件不对，可以去https://github.com/cougarlj/COMPSRA上bundle1_V1/prebuilt_db中下载
java -jar COMPSRA.jar -tk -dr -ck miRNA_hg38,piRNA_hg38,tRNA_hg38,snoRNA_hg38,snRNA_hg38,circRNA_hg38

# 下载人基因组
java -jar COMPSRA.jar -tk -dr -ck star_hg38